Obstructive sleep apnea (OSA) is a prevalent disorder of the respiratory system caused by the recurrent narrowing or collapse of the airway during sleep (1, 2). OSA is associated with male sex, higher age, overweight or obesity, and consumption of cigarettes and alcohol (3, 4). OSA is considered to induce several possible consequences, including excessive sleepiness, fatigue, metabolic disorders, and cardiovascular diseases (5–7). More importantly, subjects with OSA about to receive surgery under general anesthesia should be properly managed since OSA is associated with adverse postoperative outcomes, such as pulmonary complications, cardiovascular complications, delirium, and even mortality (8–10). Nevertheless, it is generally considered that OSA is largely underdiagnosed (11, 12). Therefore, it is critical to recognize OSA to better manage subjects about to receive surgery under general anesthesia.

Preoperative OSA is supposed to be the abnormally high occurrence of OSA the night before surgery under general anesthesia. Patients about to receive surgery under general anesthesia may have some negative emotions, such as fear of surgery and worry about the surgical outcome (13). These negative emotions could lead to the imbalance of sympathetic and parasympathetic nervous systems; while the imbalance of sympathetic and parasympathetic nervous system is recognized as a critical factor associated with OSA (14, 15). However, the information on preoperative OSA assessment is still lacking.

Cardiopulmonary coupling (CPC) is a technique generating data on heart rate variability and respiration (16). Compared with laboratory-based polysomnography (the gold standard diagnostic modality of OSA), CPC is characterized by convenience and portability (17). Several studies have suggested that CPC shows high performance for the diagnosis of OSA (18, 19). For instance, a study includes subjects undergoing full-night in-laboratory polysomnography and CPC simultaneously; the results reveal software-generated apnea-hypopnea index (AHI) by CPC and AHI data by polysomnography are highly correlated with each other; this study also revealed that the software-generated AHI by CPC show excellent performance for diagnosing OSA, especially severe OSA with sensitivity of 100% and specificity of 93.63% (18). Another study discloses that in children with severe OSA, the AHI from CPC is not different from AHI from polysomnography, suggesting the good accuracy of CPC for diagnosing severe OSA in children (19). Xie et al. (20) report that sensitivity of CPC for recognizing AHI ≥ 5/h, ≥10/h, ≥15/h, ≥20/h, and ≥30/h by polysomnography are 0.82, 0.93, 0.96, and 0.77, respectively, and the specificity are 0.50, 0.75 0.72, 0.80, and 0.86, respectively. Ma et al. (21) show that for AHI ≥ 5/h, ≥ 15/h, and ≥ 30/h by polysomnography, CPC has a sensitivity of 93.8, 92.7 and 89.5%, and specificity of 67.8, 72.2 and 79.8%, respectively.

Therefore, the current study aimed to evaluate preoperative OSA by CPC and its correlative factors in subjects who had plans to receive general anesthetic for elective surgery.

The current study revealed that the prevalence of preoperative OSA was 71.7%. We also identified that age, pNN20, LF, HF, and LF/HF ratio were independently associated with preoperative OSA. The linear regression analyses revealed that preoperative OSA showed an independent, negative correlation with pNN50 but an independent, positive correlation with LF/HF ratio.

OSA is considered to be a prevalent disorder affecting about 25% of all adults (23). Nevertheless, the detailed prevalence of OSA can vary greatly depending on the studied population, region, and screening method for OSA. For instance, a study conducted in the southern region of China uses a photoelectric reflector sensor evaluating the pulse oxygen saturation to screen OSA in 3,650 adults; the data shows that 30.7% of subjects have OSA (24). This study also uses a sleep questionnaire to screen OSA, and the prevalence of sleep questionnaire-screened OSA is 42.8% (24). A meta-analysis reviews eight studies with 11,009 subjects in India (mean age ranges from 35.5 years to 47.8 years) and concludes that the pooled prevalence of OSA is 11% in total adults, 13% in males, and 5% in females (25). Another meta-analysis reviewed 98 articles, which shows that the estimated prevalence of OSA was 54%; however, heterogeneity exists among the studies (26). Nevertheless, the information on the prevalence of preoperative OSA is not sufficient. On the other hand, patients with OSA about to receive surgery under general anesthesia may encounter difficulty in airway management and postoperative complications such as pulmonary complications, cardiovascular complications, delirium, and even mortality; meanwhile, the type and dose of anesthetics should be carefully considered since different anesthetics have different effect on respiratory system, and patients with OSA are at high risk of respiratory depression (27). Under this scenario, it is critical to evaluate preoperative OSA.

The current study utilized CPC to screen preoperative OSA, which was characterized by convenience and high screening accuracy (17–19). Another advantage of CPC is that it may have less first-night effect. The first-night effect refers to the reduced sleep quality when a subject moves to an unfamiliar environment; it may also occur when wearing complicated equipment for the first time when sleeping (28). It is supposed that the first-night effect could interfere with the diagnosis of OSA (29). Considering the equipment of CPC is more simplified compared with polysomnography, it may induce less first-night effect, thus promoting the screening of preoperative OSA. The data showed that by CPC screening, the prevalence of preoperative OSA was 71.7%, which was higher compared with previous studies (24–26). The probable reasons were supposed as follows: (1) The screening method for OSA in the current study might be different from previous studies. In the current study, CPC was used to screen OSA, which was characterized by high sensitivity and accuracy (30–32). (2) Patients about to receive surgery under general anesthesia might have negative emotions such as fear of surgery, concerns about surgical outcomes, etc., which could induce the activation of sympathetic nervous system and subsequently lead to preoperative OSA (13, 14). These data suggested that preoperative OSA was quite prevalent in patients about to receive surgery under general anesthesia, and CPC was feasible to screen preoperative OSA in these patients.

Identifying the factors associated with preoperative OSA could be meaningful to improve the management of preoperative OSA. The current study included the clinical characteristics, sleep quality characteristics, and HRV characteristics, then used forward stepwise multivariate logistic to analyze the independent factors associated with preoperative OSA. The data revealed that higher age, higher HF, higher LF/HF, lower pNN20, and lower HF were independently associated with a higher probability of preoperative OSA. The possible explanations are as follows: (1) Age is a well-recognized risk factor for OSA (1). The airway of individuals of higher age might collapse more easily to due to loss of collagen (33). (2) pNN20 was a time-domain parameter of the HRV, which might reflect the activity of parasympathetic nervous system (34, 35). LF, HF, and LF/HF ratio are frequency-domain parameters of HRV, which reflect the balance of sympathetic and parasympathetic nervous system (36). While the imbalance of sympathetic and parasympathetic nervous system is associated with OSA (37–39). Our findings also suggested that age and imbalance of sympathetic and parasympathetic nervous system were also independently associated with a higher probability of moderate-to-severe and severe preoperative OSA. These findings implied that additional perioperative care or monitoring should be given in subjects with higher age or imbalance of sympathetic and parasympathetic nervous system. The current study measured SDNN and SADNN due to the following reason: SDNN and SADNN are vital parameters in the HRV analysis, which could reflect the function of sympathetic and parasympathetic nervous system.

To further verify the association of preoperative OSA with the imbalance of sympathetic and parasympathetic nervous system, the current study conducted linear regression analyses. The data revealed that preoperative OSA showed an independent, negative correlation with pNN50 but an independent, positive correlation with LF/HF ratio. This finding further revealed the intercorrelation between preoperative OSA and the imbalance of sympathetic and parasympathetic nervous systems. Moreover, it was found that age showed an independent, negative correlation with pNN50; male, sleep efficiency, and awakening time presented independent, positive correlations with LF/HF ratio, while age and bedtime possessed independent, negative correlations with LF/HF ratio. The positive correlation of male with LF/HF ratio suggested the positive association of male with the activity of sympathetic nervous system, which was partly in accordance with previous studies (40, 41). These data further provide potential evidence for improving the management of preoperative OSA in patients about to receive surgery under general anesthesia.

The current study detected SDNN and SADNN due to their importance and application in the HRV analysis. Both SDNN and SDANN are important parameters for assessing the function of the cardiac autonomic nervous system. The autonomic nervous system is essential for maintaining the body’s physiological homeostasis and adapting to environmental changes, and HRV, as an indirect reflection of autonomic nervous system activity, can provide important information about the health and functional status of the heart. Specifically, SDNN and SDANN are able to reflect changes in heart rate over different time scales. SDNN mainly reflects the HRV over a shorter period of time, such as the change in heartbeat intervals per minute, while SDANN focuses on the average change in HRV over a longer period of time, such as 24 h. By measuring these two indicators, we can fully understand the overall characteristics and the changes in different time scales of HRV, so as to more accurately assess the functional status of the cardiac autonomic nervous system (42).

Several limitations in the current study should be clarified. First, the difference in the type of surgery might affect the findings of this study. Second, the current study excluded subjects with chronic sleep difficulty; thus, the findings of this study could not be applied to these individuals. Third, the current study did not evaluate the postoperative complications and their associations with preoperative OSA, which might be evaluated in further studies. Fourth, the current study did not assess the pre-existing OSA or the risk of OSA by STOPBANG criteria in the patients before the initiation of this study, which could be a bias. Fifth, the current study used CPC for diagnosing OSA due to its convenience and feasibility for assessing OSA before the night of surgery. While the gold standard for OSA diagnosis is using polysomnography, and our findings should be further validated. Sixth, the current study was a cross-sectional study and did not evaluate the surgical outcomes, including complications. The impact of preoperative OSA on the outcomes of surgery should be investigated in further studies.

Collectively, preoperative OSA is prevalent, and it is associated with age and imbalance of sympathetic and parasympathetic nervous system. The findings of this study suggest that CPC is helpful for screening preoperative OSA and its risk factors, which may promote the management of preoperative OSA.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding authors.

This study was approved by the Ethics Committee of Hebei Provincial Hospital of Traditional Chinese Medicine (No. HBZY2020-KY-067-02). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

SH: Conceptualization, Formal analysis, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. GZ: Conceptualization, Formal analysis, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing. XL: Formal analysis, Methodology, Validation, Writing – review & editing. CW: Formal analysis, Methodology, Validation, Writing – review & editing. JL: Investigation, Methodology, Writing – review & editing. WH: Conceptualization, Methodology, Project administration, Supervision, Writing – review & editing. LK: Conceptualization, Methodology, Project administration, Supervision, Writing – review & editing.