Salt, Diuretics, and Obstructive Sleep Apnea.

To the Editor:

We read with great interest the article by Giatti and colleagues, which was published recently in AnnalsATS (1). The authors aimed to investigate a possible involvement of sodium intake in obstructive sleep apnea (OSA) and found a significant association between sodium excretion level and OSA but only in hypertensive individuals. The most important issue to consider when interpreting the results was that the authors analyzed sodium excretion levels during the night (between 7 p.m. and 7 a.m.).

The relationship between blood pressure (BP) and diurnal variation in urinary sodium excretion has already been reported (2). BP was positively associated with nocturnal sodium excretion levels; however, no significant association was observed with BP during daytime. Therefore, in a separate analysis by BP level, hypertensive individuals excreted larger amounts of sodium during nighttime than normotensive individuals, whereas daytime excretion level was higher in normotensives (2). Higher urinary sodium excretion rate in hypertensive individuals was also observed in other studies (3, 4). Given the differences in diurnal rhythm of sodium excretion between normotensive and hypertensive individuals, it is difficult to estimate dietary salt intake per day from nighttime sodium excretion level alone, especially in hypertensive individuals.

Salt sensitivity of BP is present in some members of the population, wherein BP exhibits changes parallel to changes in salt intake (5). Salt sensitivity is a pathophysiology of increased nocturnal sodium excretion in hypertensive, or salt-sensitive, individuals. They are likely to have diminished renal sodium excretory capability and require longer time for natriuresis than non–salt-sensitive hypertensive or normotensive individuals, which results in carrying pressure natriuresis over into the night (6). This is evident from the results of an experimental epidemiological study, which showed that salt-sensitive essential hypertensive individuals had higher nocturnal BP to enhance pressure natriuresis during the night, and nocturnal BP of salt-sensitive hypertensive individuals, but not of non–salt-sensitive hypertensive individuals, further increased via salt loading during daytime (7). An involvement of salt sensitivity was also evident from the results that show that salt restriction normalizes not only daytime but also nighttime BP in essential hypertensive patients with salt sensitivity (8).

Regarding the relationship between OSA and nocturnal sodium excretion, it has been reported that the apnea–hypoxia index was positively correlated with sodium excretion during the nighttime but not during the day (5). A plausible mechanism by which OSA increases natriuresis during the night is increased intrathoracic pressure and consequent larger venous return to the atrium, which in turn enhances natriuretic peptide secretion from the atrium (9). A close positive association between the 3% oxygen desaturation index and nocturnal urination frequency was also found in our observational study in the general population (10), which is in accordance with the pathophysiology of larger sodium excretion during the night in individuals with OSA.

Given these research findings, it is possible that coexistence of hypertension and OSA synergistically increases nighttime, but not daytime, sodium excretion levels, which supports Giatti and colleagues’ findings, namely that sodium excretion during the night is associated with OSA but only in hypertensive participants. As described by the authors, fluid retention during the day and its shift into the neck by lying down at night, which causes upper airway narrowing via increasing tissue pressure, may be another reason for the association between excessive salt intake and consequent body fluid retention and OSA. However, to clarify whether high salt intake or sodium itself increases the risk of OSA, further studies comparing differences between daytime and nighttime sodium excretion and their relationship with OSA are required.