In this study, we hypothesised that proANP, proADM and copeptin levels are significantly higher in patients with OSA, when compared to patients without OSA. Our results indicated that only copeptin levels differed significantly among patients with and without OSA at baseline. We found a significant circadian variation in the levels of proADM, which increased in the evening and proANP, which decreased in the evening. In patients with OSA, the levels of proADM significantly changed after 1 and 6 months on CPAP therapy, when compared to baseline. Additionally, proANP levels significantly decreased after 12 hours on CPAP therapy, as compared to baseline levels. Those changes were significant irrespective of disease severity and the existence of symptoms.
The current findings come in partial agreement with the results of Cinarka et al, (31) who demonstrated that copeptin levels were higher in 116 patients with OSA, when compared to 27 controls. However, they also proposed that copeptin levels in OSA patients with AHI ≥ 30 were significantly higher when compared to OSA patients with AHI < 30 and that copeptin levels can be used as a predictor for severe OSA, being weakly correlated with AHI, ODI, arousal index and CRP. Our results, conversely, indicated that copeptin levels are significantly correlated with neither AHI nor ODI and that copeptin levels are not significantly different among patients with mild, moderate and severe OSA. These discrepancies might be attributed to specific patient characteristics in both studies as well as the lack of information about whether patients included in the Cinarka study were on CPAP therapy at the time when copeptin levels were measured. (31)
We found a circadian rhythm in the levels of proADM, which increased in the evening and in the levels of proANP which decreased in the evening. Indeed, in previous studies proADM and proANP demonstrated circadian variations. (32–34) However, copeptin did not present with a circadian variation, which is in line with the study of Darzy et al. (35)
Conflicting results exist about the effects of CPAP therapy on plasma proADM levels in OSA patients. In a study by Schulz et al. including 41 OSA patients and 28 controls without sleep-disordered breathing, OSA patients had markedly elevated ADM concentrations at baseline when compared to the controls. (36) After two nights of CPAP therapy, in 28 OSA patients ADM levels significantly decreased when compared to pre-CPAP values, and after 8 months of CPAP treatment, in 11 OSA patients, ADM levels further declined to levels similar with the controls. (36) On the other hand, in a small study including 15 OSA patients and 10 controls, Wolk et al. could not find any difference between ADM levels at baseline and ADM levels after 4 hours of CPAP therapy. (37) In our study, we included a larger number of patients and controls and demonstrated that there was no significant difference in proADM levels among patients with and without OSA and that in patients with OSA, proADM levels significantly changed after 1 month on CPAP therapy and significantly decreased after 6 months on CPAP therapy, when compared to baseline. Adjustments for all comorbidities, hours on CPAP and efficacy of treatment (as assessed by the AHI and the ESS on CPAP) did not change these results. ADM is secreted from various organs and is mainly produced by vascular endothelial cells, playing an important role as a vasodilator, positive inotropic, diuretic, natriuretic and bronchodilator. (38, 39) Several stimuli, such as hypoxia, shear stress and inflammatory cytokines can induce proADM production and all these stimuli are increased in patients with OSA. (40) Moreover, it has been identified as a prognostic marker able to stratify mortality risk in sepsis patients with different degrees of organ failure, (41) to predict all-cause mortality in pulmonary embolism (42) and increased complications and higher mortality rates in patients suffering from community acquired pneumonia. (43) OSA-induced hypoxic stress and oxidative stress increase circulating inflammatory mediators, including adhesion molecules, inflammatory cytokines and C-reactive protein, leading to hypertension and cardiovascular events. (44, 45) This stress and the related inflammatory molecules are implicated in the production of adrenomedullin which is a potent vasodilator. (46) Adrenomedullin has also been associated with the magnitude of oxyhemoglobin desaturation in OSA patients and with production of reactive oxygen species by leukocytes and treatment with nasal CPAP reduced these parameters in patients with OSA. (47) Therefore, it seems that the upregulation of proADM in OSA constitutes an adaptive counteractive mechanism to protect against cardiovascular diseases in patients with OSA. The reversal of the pathophysiological mechanisms of OSA by implementing non-invasive ventilation might in turn explain the longitudinal decrease of ProADM observed in patients on CPAP therapy.
Previous studies investigating ANP levels in OSA demonstrated that treatment with CPAP acutely reduces plasma ANP levels. (34, 48) We also demonstrated that in 82 patients with OSA, proANP levels significantly decreased after 12 hours on CPAP therapy, as compared to baseline levels. Those changes were significant irrespective of disease severity and the existence of symptoms. That comes, nonetheless, in contrast with the results of Mackay et al, who demonstrated that in 9 patients with OSAS, treatment with CPAP for 2 days did not decrease plasma ANP levels. (49) Nevertheless, the study of Mackay et al. included a very small number of patients, when compared to our study. Other studies, could not find any association between OSA and plasma ANP levels (50–52) or even found inverse relations between plasma ANP and AHI. (49) Similarly, in our study, proANP levels were not significantly different between patients with and without OSA, and furthermore, proANP levels were not significantly associated with the severity of OSA, as assessed by the AHI.
A potential limitation of our study is the relatively short follow-up period. A longer term of treatment would have provided a better insight to the effect of CPAP on the prohormone levels. However, compared to previous studies, (36, 37, 49–52) this follow up period was the longest. Moreover, the sample size was estimated based on changes in copeptin levels and thus a beta error for proADM and proANP levels cannot be excluded. Nevertheless, no other study so far has investigated proADM and proANP level changes in such a large group of OSA patients. Another limitation of our study is that we utilized only standardised diagnostic and therapeutic methods for OSA and that prohormone levels were measured only in five different time points and not hourly. Yet, to our knowledge, this is the first study that has investigated proADM, proANP and copeptin levels in various time points in a large group of fully characterised OSA patients.