Obstructive sleep apnea (OSA) is characterized with recurring episodes of partial or complete collapse of the upper airway during sleep, resulting in fragmented sleep, intermittent hypoxia, intrathoracic negative pressure, and other pathological physiological processes which are especially associated with cardiovascular complications [1], such as hypertension [2], myocardial ischemia, heart failure [3], arrhythmia, stroke [4], and even sudden cardiac death [5]. The prevalence of OSA in adults in the general population ranges from 9% to 38% [6], and even higher in an aging population, what’s more, the incidence of OSA varying from 20% to 60% in the people over 65 years and increased to 78% in women and 90% in men aged 70 to 85 years [7]. In addition, previous study has demonstrated that OSA, asthma, gastroesophageal reflux disorder (GERD), and obesity have shared inflammatory pathways; OSA prevalence was also greater among those with asthma patients. The rate of OSA in subjects with asthma at 49.5% [8], in GERD populations was found to be 65% [9]. According to epidemiologic data, the higher prevalence of OSA also occurs in obesity [10].

Currently, continuous positive airway pressure (CPAP) is the optimal therapy for moderate to severe OSA, especially for those without anatomical obstruction [11]. It stabilizes the airway to prevent collapse, improve gas exchange, relieves sleep fragmentation and the chronic intermittent hypoxemia caused by OSA [12] and also improves endothelial cell dysfunction [13]. These improvements benefit OSA symptoms and possibly CV outcomes in some group but not all patients group [14, 15], the effects of CPAP treatment still remain debated.

It is well-known that expansion sphincter pharyngoplasty (ESP) is an effective upper airway surgery choice for the treatment of OSA, successful ESP associated with a reduction in sympathetic activity of the heart, which might be connected with lower incidence of cardiovascular disease [16]. Some studies suggest that CPAP therapy can effectively reduce the risk of cardiovascular (CV) events in OSA patients with previous cardiovascular or cerebrovascular disease [17, 18, 19, 20]. However, recent randomized control trials (RCTs) failed to show the aforementioned benefits [21, 22, 23]. We conducted this meta-analysis of RCTs by synthesizing existing RCTs to investigate the effect of CPAP therapy on prevention of recurrent cardiovascular or cerebral events in OSA patients with history of CV diseases.

To our knowledge, the present meta-analysis is the first one to focus on assessing the role of CPAP therapy on MACCEs among participants with OSA and previous cardiovascular/cerebrovascular disease. In this systematic review and meta-analysis of current data from 6 RCTs, the key findings were as follows: (1) CPAP therapy failed to significantly reduce the risk of MACCEs or other outcomes, such as myocardial infarction, atrial fibrillation, heart failure, all-cause death, angina, revascularization compared with usual treatment. (2) Subgroup analyses confirmed that adequate use of CPAP ($>$4 hours) is associated with statistically significant and clinically meaningful improvement in CV mortality and stoke. A previous study has shown that OSA has the most profound adverse effects on stroke and the cerebral circulation [30]. Our findings further demonstrate the importance of adequate CPAP treatment for OSA patients.

A previous review was published by Khan et al. [31] that included 7 RCTs (including the SAVE study), which demonstrate that CPAP therapy might reduce MACE when usage time is more than 4 hours/night. Our study is inconsistent with theirs which maybe dues to the fact that this review pooled studies whose participants without previous CV history, and other different interventions such as nocturnal supplemental oxygen, healthy lifestyle and sleep education were used. In our review, five RCTs reported the CV mortality, which showed evidence of an association between CPAP and risk of CV mortality (OR 0.25, 95% CI [0.08–0.77], p = 0.02). Labarca et al.’s [32] meta-analysis including 8 RCTs studies aimed to assess the benefits of CPAP for secondary prevention in participants with high cardiac risk and previous cardiovascular outcomes, but were unable to show any benefits of CPAP on cardiovascular outcomes. However, they raised concerns about bias, CPAP adherence, and the differing populations included in each RCT which may have reduced the strength of the findings. In patients with a high cardiac risk and previous CV outcomes, pooled data have failed to show an evidence of efficacy of CPAP, and subgroup analysis showed no significant reduction in the risk of cardiovascular mortality and stroke. The results of our subgroup analysis differ from this study which maybe attributed to different participants as our review only included participants with previous cardiovascular events and OSA. Previous studies suggested that CPAP therapy is associated with a slight decrease in blood pressure in non-sleepy patients with OSA [33, 34]. Otherwise, the use of CPAP in OSA with excessive daytime sleepiness, most likely obtain potential cardiovascular benefit [35]. There are only 6 RCT in the final analyses, most studies included patients with non-sleepy or exclude ESS scores over 15 points, thus, further research should collect data on the efficacy of CPAP in pre-specified populations and focus on populations at risk for cardiovascular events. Labarca et al.’s [32] review included 3 RCTs which reported primary prevention and others reported secondary prevention of cardiovascular disease. In addition, the SAVE and ISAACC trial occupies a large weight in this analysis, which may have an influence on the results. Besides, Labarca et al.’s [32] review included participants who use sham CPAP as control group. As discussed before, there is no true sham CPAP or other placebo for CPAP in long-term studies in OSA patients with cardiovascular/cerebral disease [36]. Furthermore, not everyone can tolerate this therapy because it requires them to sleep with a nasal or full-face mask that is connected by a tube to a machine. Sham positive pressure ventilation may worsen sleep disturbances because of mask coverage and without effective pressure [37]. CPAP adherence is an important factor, and benefits might be underestimated related to lower CPAP compliance. Wang et al.’s [38] meta-analysis including 7 observational studies and 2 RCTs studies, which suggested that the benefits of CPAP on prevent subsequent cardiovascular outcomes in patients with cardiovascular disease and OSA, was not verified in RCTs. The result of this study maybe underpowered due to included RCTs studies is not enough.

Although our findings suggest that patient with OSA and a history of cardiovascular disease did not obtain benefits from CPAP therapy in regard to improving either MACCEs or preventing clinically relevant outcomes such as myocardial infraction, a non-significant trend in favor of revealed that CPAP therapy reduce MACEs with adequate time ($>$4 hours/night) (OR 0.77, 95% CI [0.5–1.18], p = 0.23). At the same time, favorable trend with a better compliance time ($>$4 hours/night) was noticed in cardiac mortality subgroup analysis (OR 0.25, 95% CI [0.08–0.77], p = 0.02). Analysis suggested that adequate usage time of CPAP can protect patients from cardiovascular or cerebrovascular death. In other words, CPAP therapy can minimize the harmfulness of OSA complications to a certain extent.

Our findings of CPAP treatment can improve stroke in OSA with previous cardiovascular has clinical implications. On a global scale, stroke ranks as the second cause of death [39]. A growing body of evidence suggests a clear association between OSA and the incidence of stroke [40, 41]. Patients who suffered a stroke have poor quality of life and increased mortality. There are some reasons which could explain this therapy response: when autoregulation is impaired by OSA cycles, cerebral vascular bed fluctuation exposed to OSA-related increase in blood pressure, and chronic intermittent hypoxemia [42]. Therefore, the use of CPAP in OSA patients who suffer from stroke may be necessary. Our findings emphasize the importance of CPAP usage in OSA patients with cerebral diseases, which could potentially prevent adverse cerebrovascular consequences. What’s more, in clinical practice, health care providers should emphasize the importance of CPAP usage time when caring for OSA patients with stroke. Due to intermittent hypoxemia is regarded as a potential contributing factor to the pathogenesis of OSA-related comorbidities, the frequency of hypoxia maybe influents the effect of CPAP, therefore, more studies are needed to explore the differential effects of both short-term and long term hypoxemia about the effect of CPAP treatment.

There are some limitations in this analysis. First, due to the small number of studies our findings should be interpreted with caution. Second, follow-up times for the included studies were variable (range from 6 months to 84 months), and in one study follow-up was for 6 months only, which was relatively short for a cardiovascular event at follow-up. Lastly, RCTs compared CPAP therapy with usual care rather than a comparison with sham CPAP, which result in a high risk bias of blinding of participants and personnel (A true sham as pointed above is difficult with CPAP). In addition, the studies included in this meta-analysis used different analyses (adjusted, unadjusted, and propensity matched analysis) which could be a source of potential heterogeneity.

Our meta-analysis suggests that if CPAP is used for more than four hours there is a decrease trend of the incidence of MACCEs, especially, we can observe that increasing CPAP compliance time can have positive effect on cardiac mortality. Considering the risk of bias in the included studies, the usage time of CPAP, the duration of follow-up, and the included population, the credibility of the results may be reduced, and future studies need to improve the quality of CPAP use to further confirm these results.

AF, atrial fibrillation; CI, confidence interval; CPAP, continue positive airway pressure; CV, cardiovascular; GERD, gastroesophageal reflux disorder; ESP, expansion sphincter pharyngoplasty; HF, heart failure; MACCEs, Major Adverse Cardiovascular or Cerebrovascular Endpoints; MI, myocardial infarction; OR, odds ratio; OSA, obstructive sleep apnea; PG, cardiorespiratory polygraph; PSG, polysomnography; RCT, randomized controlled trials; TIA, transient ischemic attack.

HL—Guarantor of the paper, manuscript conception, literature search, data extraction, data analysis, manuscript editing, and final approval read; YYP—Data extraction manuscript editing, and final approval read; YKL—Data analysis, editing the English text of a draft of this manuscript; FF and JES—Guarantor of the paper, manuscript conception, final approval read, editing the English text of a draft of this manuscript; YJZ and LRY—Literature search, Data extraction, data analysis, manuscript editing. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Not applicable.

All the authors appreciated the working staff of sleep medicine center.

This work was supported by the National Key Research and Development Program of China (No. 2020YFC2003600).

The authors declare no conflict of interest.