Blood pressure dipping during REM and non-REM sleep in patients with moderate to severe obstructive sleep apnea

A limited number of papers have addressed the association between non-dipping-blood pressure (BP) obstructive sleep apnea (OSA), and no study has assessed BP-dipping during rapid eye movement (REM) and non-REM sleep in OSA patients. This study sought to noninvasively assess BP-dipping during REM and non-REM (NREM)-sleep using a beat-by-beat measurement method (pulse-transit-time (PTT)). Thirty consecutive OSA patients (men = 50%) who had not been treated for OSA before and who had > 20-min of REM-sleep were included. During sleep, BP was indirectly determined via PTT. Patients were divided into dippers and non-dippers based on the average systolic-BP during REM and NREM-sleep. The studied group had a a median age of 50 (42–58.5) years and a body mass index of 33.8 (27.6–37.5) kg/m2. The median AHI of the study group was 32.6 (20.1–58.1) events/h (range: 7–124), and 89% of them had moderate-to-severe OSA. The prevalence of non-dippers during REM-sleep was 93.3%, and during NREM-sleep was 80%. During NREM sleep, non-dippers had a higher waist circumference and waist-hip-ratio, higher severity of OSA, longer-time spent with oxygen saturation < 90%, and a higher mean duration of apnea during REM and NREM-sleep. Severe OSA (AHI ≥ 30) was defined as an independent predictor of non-dipping BP during NREM sleep (OR = 19.5, CI: [1.299–292.75], p-value = 0.03). This short report demonstrated that BP-dipping occurs during REM and NREM-sleep in patients with moderate-to-severe OSA. There was a trend of more severe OSA among the non-dippers during NREM-sleep, and severe OSA was independently correlated with BP non-dipping during NREM sleep.


Materials and methods
Subjects. We recruited consecutive patients who had not used positive airway pressure therapy for OSA from the sleep disorders clinic. Inclusion criteria were ≥ 18-years-old, OSA diagnosis based on overnight polysomnography (PSG) (SOMNOmedics GmbH, Randersacker, Germany), and having REM-sleep (≥ 20-min) during monitoring between April and October 2018. Exclusion criteria included daytime hypercapnia, alternative treatments for OSA (e.g., surgical procedures), known neuromuscular, respiratory, or cardiovascular diseases, secondary causes of hypertension (other than OSA), or requirement for home oxygen therapy. All patients underwent arterial blood gas analysis.
Ethics approval was obtained from the Institutional Review Board in our institute, and all participants signed a written informed consent form.
Blood pressure measurement. The recruited patients were instructed to stop smoking on the evaluation day. None of the participants drink alcohol. During sleep, BP was indirectly determined via a pulse-transit-time (PTT), which is validated-method using the DOMINO-Software (DOMINO-2.2.0 provided with the SOMNOscreen-plus, Randersacker, Germany) 18 . We have described the details of the used method in a previous paper 19 . In brief, the noninvasive measurement of BP using the SOMNOscreen plus has been validated following the European Society of Hypertension protocol 18 . The system assesses BP continuously and noninvasively (beat-tobeat determination of PTT), calculated as the interval between the ECG R-waves and the detection of the corresponding pulse wave (revealed from the finger photoplethysmography signal) at the peripheral site.
The algorithm utilizes the relationship between BP and the pulse-wave-velocity (PWV), and good results have been reported for BP measurement using the PTT in OSA patients 20 . The computation of the systolic blood pressure (SBP) depends on a non-linear correlation between BP and PTT 13 . Calibration was done based on a PWV-BP relation following the manufacturer's instructions 18 . The mean values of SBP during REM and NREMsleep were used in the analysis.
Awake-BP was determined via measuring BP for one hour while awake in the sitting position using the same (PTT) method between 9 and 10 AM on the day of the study, and the mean awake-BP was also used in the analysis. The all-night PTT recording was viewed, and data segments containing movement artifacts were excluded from the analysis. Clean 30 s epochs were analyzed, and the mean SBP and DBP were calculated. BP determined automatically beat-to-beat with the DOMINO software based on a non-linear pulse wave velocity-SBP function in combination with an initial BP calibration.
The patients were divided into dippers if a reduction in the mean systolic-BP of ≥ 10% of the wake systolic-BP was documented during sleep; they were called non-dippers if the reduction was < 10%. Further, patients were divided into dippers and non-dippers based on the average systolic-BP during REM and NREM-sleep.
Polysomnography. All patients underwent a standard overnight attended type-I PSG (SOMNOscreenplus, Randersacker, Germany). PSG scoring was done manually following the American Academy of Sleep Medicine scoring criteria by a certified sleep technologist who had no access to the clinical information 21 . Apnea was scored when there was a drop of flow signal of ≤ 90% for at least 10-s. Hypopnea was defined as a drop in flow signal by ≥ 30% for at least ≥ 10 s, associated with arousal or ≥ 3% oxyhemoglobin desaturation. The apnea-hypopnea index (AHI) was calculated during REM and NREM sleep. The severity of OSA was graded according to the AHI: 5-< 15, mild OSA; 15-< 30, moderate OSA; and ≥ 30, severe OSA 22 .
Statistical analysis. For categorical variables, data were expressed as numbers (percentages), and for continuous variables, mean and standard deviation were used and median with interquartile range (IQR) in the Tables. The Student t-test was used to compare continuous variables if data passed the normality test (Kolmogorov-Smirnov); upon failing normality testing, the Mann-Whitney U test was used. For dichotomous variables, the Chi-square test was used. However, when the expected frequencies were < 5, we used the Fisher's exact test.
Multiple logistic regression analysis (Forward: Wald method) was performed to assess the association between markers of OSA severity (AHI, desaturation, apnea duration, and arousals) and BP-dipping while adjusting for age, sex, BMI, and smoking status expressed as odds ratio [OR] and confidence interval [CI]. A p-value of ≤ 0.05 was considered significant. Data were analyzed using the SPSS-statistical (version-24; Chicago, IL, USA) was used for the analyses.
Informed consent. Written informed consent was obtained from all participants. www.nature.com/scientificreports/ Research involving human subjects. The study protocol was approved by the institutional review board at King Saud University, and informed consent was obtained from all the participants prior to inclusion in this study. The used methods were carried out in accordance with the Declaration of Helsinki and the relevant guidelines and regulations.

Results
Thirty patients (males = 14) who had a median age of 50 (42-58.5) years and a BMI of 33.8 (27.6-37.5) kg/m 2 , met the inclusion criteria and were included. Nine patients have been previously diagnosed with hypertension. The median systolic and diastolic-BP of the study group were 128 (114-145) mmHg and 74 (68-85) mmHg, respectively. Among the whole group, 89% of the sample had moderate-to-severe OSA. The median AHI of the study group was 32.6 (20.1-58.1) events/h (range: 7-124). The AHI during NREM-sleep (AHI-NREM) was 30.9 (17.2-56.7) events/h, and the AHI REM-sleep (AHI-REM) was 48.4 (29-74.10 events/h. The prevalence of non-dippers during REM-sleep was 93.3%, and during NREM-sleep was 80%. Two patients had BP-dipping during REM and NREM-sleep, and four patients had BP-dipping only in NREM-sleep. Figure 1A shows an example of a patient who had BP-dipping in REM and NREM-sleep, and Fig. 1B shows an example of reverse dipping during REM-sleep. Table 1 presents demographics, comorbidities, arterial blood gases, and polysomnographic findings in dippers and non-dippers during REM and NREM sleep. The small sample size of dippers during REM sleep (n = 2) did not allow performing a comparison between dippers and non-dippers. However, during NREM sleep, non-dippers had a higher waist circumference and waist-hip-ratio, higher severity of OSA, longer-time spent with oxygen saturation < 90% (SpO 2 ), and higher mean duration of apnea during REM and NREM-sleep.

Discussion
This preliminary report demonstrated that BP-dipping is low among patients with moderate-to-severe OSA. However, the central message of this preliminary report is that BP-dipping may occur in both REM and NREMsleep in patients with moderate-to-severe OSA; nevertheless, as we hypothesized, the prevalence of BP-dipping was more during NREM than REM-sleep in OSA patients. OSA severity was apparent among non-dippers during NREM sleep. The small sample size of non-dippers in REM sleep did not allow proper comparison and detection of potential differences between dippers and non-dippers. Non-dippers during NREM had a lower percentage of N3 and a higher prevalence of diabetes mellitus than non-dippers during REM sleep.
Nocturnal BP-dipping is an active process controlled by the central nervous system; it is an essential process for the regulation of daytime-BP. In healthy individuals, as sleep progresses from N1 to N3-sleep, BP slowly declines due to increased vagal tone and decreased sympathetic-activity, and BP reaches its nadir during stage-N3 23 . On the other hand, REM-sleep is associated with intermittent increases in sympathetic-activity and BP 23 . However, findings in healthy individuals may not apply to patients with OSA. Therefore, it is essential to objectively monitor the decline in the average BP in OSA patients during REM and NREM-sleep. Changes in BP during sleep stages and in OSA patients are quick with significant swings; therefore, a method that measures these rapid BP changes in different sleep stages with accuracy is needed. A strong point of this study is the use of a valid beat-by-beat measurement of BP, which allowed us to detect the dipping in BP in Rem and NREM with accuracy. The PTT and PWV method is an accurate application of beat-to-beat BP recordings 13,15,16,24 . The beat-by-beat measurement of BP permits precise recording of the very short-term variability in BP that may accompany changes in sleep stages and obstructive respiratory events 25 .
Theoretically, sleep stages may impact nocturnal-BP changes in OSA patients through the severity of the obstructive events, duration, and accompanied arousals and desaturations in each stage. In the current study, stage-N3% was significantly lower in the non-dippers during NREM sleep. Experimental studies that assessed slow-wave-sleep deprivation on BP in healthy-volunteers suggested an impact of slow-wave-sleep on BP-dipping during sleep. Tasali et al. reported a decrease in vagal-tone and an increase in sympathetic-activity with slowwave-sleep deprivation 26 . Another study randomly deprived 11 healthy subjects of slow-wave-sleep via acoustic stimulation for one-night, and the effects were compared with one-night of undisturbed sleep 27 . Suppression of slow-wave-sleep resulted in a significant decrease in BP-dipping; however, no significant changes were reported in the morning-BP and no changes in urine catecholamine levels 27 .
There was a clear trend towards a more severe OSA among non-dippers, and severe OSA was an independent correlate of non-dipping BP during NREM-sleep. This concurs with previous studies, which demonstrated that non-dipping correlated with the severity of respiratory events in OSA patients 25 . Additionally, apnea duration and time-spent with SpO 2 < 90% were longer in the non-dippers in NREM-sleep. Previous studies demonstrated that increased duration of the obstructive respiratory event during sleep is associated with increased nocturnal BP 28,29 . Additionally, longer duration of apnea results in a higher level of hypercapnia and hypoxemia 28 , which work together to enhance sympathetic nerve activity and hence increase BP 30 .
This current short report has some strengths; it included patients with > 20-min of REM sleep to assess the changes in BP during REM-sleep, and the study used a validated beat-to-beat measurement method. Nevertheless, the study's limitations include being a preliminary report with a relatively small sample; hence, a larger www.nature.com/scientificreports/ study is needed to confirm the current findings and identify the predictors of BP-dipping during REM-sleep. Second, the present report did not include controls. Third, antihypertensive medications were not stopped in patients with hypertension. Finally, the difference between diurnal and nocturnal BP values is ideally assessed via measuring the 24-h BP recording to evaluate the changes in mean BP from day to night. However, as we used the PTT method built-in the PSG device, we could not do 24-h monitoring, as we were keen to use the same measuring method and device to avoid discrepancy between methods. To mitigate that, we measured BP using  www.nature.com/scientificreports/ the same PTT method on the day of the study for one hour between 9 and 10 AM. In normal individuals, BP usually shows two peaks, one in the morning and the other in the late afternoon or early evening 31 . Studies have reported differences in peak BP timing, while a higher morning peak was reported mainly in Asian individuals, Table 2. A comparison of demographics, comorbidities, and polysomnographic findings among non-dipper in REM and NREM sleep. PSG polysomnography, OSA obstructive sleep apnea, BMI body mass index, REM rapid eye movement, NREM non-rapid eye movement, SpO 2 arterial oxygen saturation.

Variable n = 30
Non-dippers median (Interquartile range (IQR)* www.nature.com/scientificreports/ higher evening peak was reported in other studies, mainly in European individuals 25,32 . Moreover, 24-h changes in BP can be revealed by BP measurements taken at different times of the day, e.g., morning and evening, and do not need to be a continuous measurement 25 8 . Therefore, future research should determine whether the definition of BP dipping during REM and NREM sleep should be the same.

REM Sleep (n = 28) NREM Sleep (n = 24) p-value
In the current literature and published guidelines, it is unclear whether dipping phenomena should be based on SBP and/or DBP, and there is debate among experts 33,34 . It is possible that the use of SBP is better as people get old because data collected in several studies demonstrated that DBP becomes lower in most patients when they get older 33 . Previous studies have used SBP, DBP, or MABP. According to experts in BP monitoring, this area is still ambiguous [33][34][35] . Nevertheless, the extent of percentage of nocturnal reduction is on average greater for DBP than for SBP 36 . It has also been suggested that the SBP night-to-day ratio may be similar in auscultatory and oscillometric recordings, whereas this could not be the case for the DBP night-to-day ratio 37 . Therefore, we opted to use SBP in this paper. Nonetheless, more future research is needed to elucidate this ambiguity.
In summary, this preliminary report demonstrated that BP-dipping occurs during both REM and NREMsleep in a small percentage of patients with OSA. There was a trend of more severe OSA among the non-dippers during NREM-sleep.

Data availability
Data are available upon request but need an institutional approval. www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.