Arterial bicarbonate is associated with hypoxic burden and uncontrolled hypertension in obstructive sleep apnea - The ESADA cohort
Ding Zou, Ludger Grote, Özen K. Başoğlu, Johan Verbraecken, Sophia Schiza, Paweł Śliwiński, Paschalis Steiropoulos, Carolina Lombardi, Holger Hein, Jean‐Louis Pépin, Gianfranco Parati, Walter T. McNicholas, Jan Hedner, Paschalis Steiropoulos, Johan Verbraecken, E. Petiet, Georgia Trakada, Ingo Fietze, Thomas Penzel, Ondřej Ludka, Izolde Bouloukaki, Sophia Schiza, Walter T. McNicholas, Stephen J. Ryan, R.L. Riha, John A. Kvamme, Ludger Grote, Jan Hedner, D. Zou, Dirk Pevernagie, Sébastien Bailly, Jean‐Louis Pépin, Renaud Tamisier, Holger Hein, Özen K. Başoğlu, Mehmet Sezai Taşbakan, Jitka Bušková, Pavol Joppa, Richard Staats, Dries Testelmans, Haralampos Gouveris, K. Ludwig, Carolina Lombardi, Gianfranco Parati, Maria R. Bonsignore, Francesco Fanfulla, Michael Drummond, Mafalda van Zeller, Winfried Randerath, Marcel Treml, Z. Dogas, R. Pecotic, Athanasia Pataka, Ștefan Mihăicuță, Ulla Anttalainen, Tarja Saaresranta, Paweł Śliwiński
Abstract
Blood bicarbonate concentration plays an important role for obstructive sleep apnea (OSA) patients to maintain acid-base balance. We investigated the association between arterial standard bicarbonate ([HCO3-]) and nocturnal hypoxia as well as comorbid hypertension in OSA. A cross-sectional analysis of 3329 patients in the European Sleep Apnea Database (ESADA) was performed. Arterial blood gas analysis and lung function test were performed in conjunction with polysomnographic sleep studies. The 4% oxygen desaturation index (ODI), mean and minimum oxygen saturation (SpO2), and percentage of time with SpO2 below 90% (T90%) were used to reflect nocturnal hypoxic burden. Arterial hypertension was defined as a physician diagnosis of hypertension with ongoing antihypertensive medication. Hypertensive patients with SBP/DBP below or above 140/90 mmHg were classified as controlled-, uncontrolled hypertension, respectively. The [HCO3-] level was normal in most patients (average 24.0 ± 2.5 mmol/L). ODI, T90% increased whereas mean and minimum SpO2 decreased across [HCO3-] tertiles (ANOVA, p = 0.030, <0.001, <0.001, and <0.001, respectively). [HCO3-] was independently associated with ODI, mean SpO2, minimum SpO2, and T90% after adjusting for confounders (β value [95%CI]: 1.21 [0.88–1.54], −0.16 [-0.20 to −0.11], −0.51 [-0.64 to −0.37], 1.76 [1.48–2.04], respectively, all p < 0.001). 1 mmol/L elevation of [HCO3-] was associated with a 4% increased odds of uncontrolled hypertension (OR: 1.04 [1.01–1.08], p = 0.013). We first demonstrated an independent association between [HCO3-] and nocturnal hypoxic burden as well as uncontrolled hypertension in OSA patients. Bicarbonate levels as an adjunctive measure provide insight into the pathophysiology of hypertension in OSA.