Mild increases in erythrocyte measures found in association with obstructive sleep apnea

Citation:

Li T, Tang X, Covassin N. Mild increases in erythrocyte measures found in association with obstructive sleep apnea. J Clin Sleep Med. 2021;17(7):1515–1516.

We are pleased that Li and Wang have taken interest in our study¹ and we thank them for their letter.² Although we believe that the main findings have been adequately interpreted, including addressing the potential confounding role of age, we are grateful for the opportunity to further discuss our findings and address the concerns raised in the letter.

Li and Wang call into question the clinical significance of the mild increases in erythrocyte measures we found in association with obstructive sleep apnea (OSA) severity, specifically in women. We acknowledge that, albeit elevated, the observed values were largely within normal ranges, even in older women with severe OSA. Indeed, only 4.3% of our entire study population met the criteria for polycythemia. However, it should be noted that, in women, 4 out of 6 total cases were in patients with severe OSA—a significant effect. We also recognize that, when considering the age range of our patients, the average values of erythrocyte measures we observed were remarkably consistent with the values that would be expected for their age, as reported in a previous study on blood count reference intervals in the Han population cited by the authors.³ As patients with more severe OSA were older than those with milder disease, the concern that our results may merely reflect the effect of age on erythropoiesis is justified. As noted by Li and Wang, in order to take into account confounding effects, we tested the relation between OSA severity and erythrocyte measures in multiple regression models and statistically controlled for age. Results showed that the association between OSA and red blood cell and hematocrit remained significant after adjustment in women. Additionally, results from a secondary analysis on a subset of male and female patients matched for age, body mass index, and apnea-hypopnea index further corroborated these findings. Although we cannot exclude residual confounding that may persist even after statistical correction, and we acknowledge that dissecting the pathophysiological contribution of individual risk factors remains challenging despite using adequate statistical methods, collectively our results point strongly toward an independent, sex-specific effect of OSA on erythrocyte measures.

It is well known that the menopause transition is associated with a striking increase in chronic and even acute illnesses in women.⁴ It is pertinent to note that, in our study, we also found that the relationship between hematologic metrics and cardiometabolic risk markers was more pronounced in women than in men, hence further emphasizing the clinical relevance of our findings. Based on our data, we believe that OSA may play a significant role in erythropoiesis and related adverse health implications and be especially important for women’s health.

DISCLOSURE STATEMENT

All authors have seen and approved the manuscript. Work for this letter was performed at the Sleep Medicine Center, West China Hospital, Sichuan University, China. This work was funded by the National Natural Science Foundation of China (81530002). The authors report no conflicts of interest.