In this study, a screening instrument was developed for primary screening of OSA in ACS patients. Our study found that waist circumference, micrognathia combined with SBQ improved the screening power of OSA in ACS patients, and its utility as a screening tool was confirmed by strong specificity, PPV, NPV, PLR and NLR. Furthermore, we established a simple and easy-to-use nomogram model which is integrated the 3 predictors to predict the risk of moderate/severe OSA in ACS patients. The establishment of this screening tool can refer high-risk patients with moderate/severe OSA to sleep center to relieve pressure on existing healthcare resources.
Despite the high prevalence of OSA in patients with ACS, OSA is often underdiagnosed and undertreated in cardiovascular practice [15]. Therefore, it is particularly important to develop strategies for better screening for sleep apnea in newly admitted ACS patients. Current screening strategies, such as the ESS and the SBQ, rely primarily on self-reported patient symptoms and provide mixed results. The sleep questionnaire alone has low diagnostic accuracy and is not recommended as the only diagnostic test [12, 16, 17]. The ESS is used frequently to measure excessive daytime sleepiness in research and clinical settings. We found that ESS scores were not related to moderate/severe OSA in the multivariate logistic regression model, so ESS could not predict OSA prevalence in ACS patients. Kendzerska et al. showed that ESS can be recommended for group-level comparisons, but not for individual-level comparisons due to its internal consistency [18]. ESS should not be used to evaluate the effectiveness of therapeutic interventions or to prioritize access to services [19]. The SBQ is a more reliable and easy-to-use screening tool that can stratify patients at risk for OSA based on their score. Moreover, the study by Hwang et al. showed that the SBQ could be an effective tool for screening OSA (AHI ≥ 5) in patients with cardiovascular risk factors [20]. Although the SBQ has been validated in different populations, there may still be potential biases in some validation studies due to the possible self-selection bias of patients themselves [11]. In veterans, the SBQ alone is not sufficient to confirm the presence of severe sleep apnea [21]. The SBQ also cannot rule out the presence of OSA in patients with chronic kidney disease and end-stage renal disease [22]. Consequently, to ensure reliably screening, it is recommended to validate the SBQ in a specific target population and incorporate objective indicators [23].
Previous research has shown that anthropometric characteristics, such as neck circumference, can be used as predictors of OSA severity [7]. Kim et al. used cephalometric and other variables to develop a formula to predict Koreans with suspected OSA [24]. In addition, regional obesity has also been shown to correlate with the severity of OSA, with neck fat having a direct effect on upper airway patency in women and abdominal obesity being the predominant factor in men [25]. Among middle-aged men, waist-to-hip ratio strongly predicted sleep disordered breathing in obese and nonobese men [6]. OSA is a multifactorial disease, airway collapse and poor pharyngeal muscle reactivity are one of the main pathophysiological factors [26]. Oropharyngeal crowding is a local anatomic factor, the more crowded the upper airway, the more severe the OSA. Oropharyngeal exercises can increase muscle tone, endurance, and coordination of movements in the pharyngeal and peripharyngeal muscles. Studies have shown that oropharyngeal exercises can significantly improve OSA [27]. Therefore, patient oropharyngeal parameters, such as micrognathia, upper airway, and antiadoncus, which are easily measured by nurses, should also be considered for analysis of whether they are associated with OSA severity [28].
Previous studies have shown that the visceral adiposity index (VAI) is a better predictor of clinical and coronary angiographic severity assessment in patients with ACS than other obesity indices [29]. VAI is determined by WC, BMI, fasting triglycerides (TG), and high-density lipoprotein cholesterol [30]. Compared with peripheral adiposity (BMI), central adiposity (WC) was more predictive in ACS patients [31]. Consistent with our research, among numerous nurse-led anthropometric parameters, only WC (OR 1.075, 95% CI 1.044–1.108, P < 0.001) was independently associated with the prevalence of moderate/severe OSA. Compared with no/mild OSA patients, moderate/severe OSA patients had significantly larger waist circumferences (P < 0.001). In addition, studies have shown that abdominal obesity, but not general obesity, seems to play a more important role in OSA [32]. Waist circumference is associated with the severity of OSA. Tom et al. suggested that waist measurements were more correlated with specific disease severity (SaO2 minimum and AHI) than BMI in OSA subjects [33]. These findings provide an easily measurable adjunct to assessing OSA risk. Craniofacial anomalies are common in patients with moderate/severe OSA. Both midface and mandibular hypoplasia contribute to OSA in these populations [34]. For this reason, it is important to identify these features as soon as possible. In our study, micrognathia was significantly associated with the prevalence of moderate/severe OSA (OR 2.019, 95% CI 1.020–4.000, P = 0.044). The study by cielo et al. showed that micrognathia was associated with more significant OSA than normal infants and that OSA improved in most infants with micrognathia after surgical correction [35].
Sleep questionnaires such as ESS and the SBQ may have recall biases, and patients may have unclear or no knowledge of their sleepiness and apnea [36]. These factors greatly decrease specificity and highlight the need for a combined-modality screening tool. The combined-modality screening tool in our study differs from these other screening tests in a number of ways. Firstly, it is specifically designed to screen moderate/severe OSA in patients with ACS. Population heterogeneity is a very important consideration, and studies have shown that arterial hypertension and neck circumference are important variables in patients with severe OSA living at high altitudes[8]. Sole et al found that traditional OSA predictors (eg, gender, Epworth score) performed poorly in more advanced COPD patients [37]. Secondly, this tool combined physical examination characteristics and self-reported symptoms to increase sensitivity, especially in the case of atypical OSA manifestations. Friedman et al. also used a screening test combining symptoms and objective results to diagnose OSA [38]. In contrast, our tool uses a single nomogram to screen moderate/severe OSA in patients with ACS. A nomogram is simple to calculate and can be used to quickly rule out low-risk OSA [39].
If mild/no OSA patients are referred to the sleep centers for PSG can further strain existing resources, resulting in a waste of money and time. Considering the heterogeneity of patients and the need for a broader screening strategy for OSA, it is imperative to develop simpler and more reliable screening modalities to prioritize PSG treatment [40]. Our study now develops a new screening strategy that combines anthropometric and oropharyngeal measurements with SBQ, which take into account anatomical criterion. Compared with the traditional use of sleep questionnaires (ESS or SBQ) alone for screening admitted patients, the diagnostic accuracy of the new screening model (WC, micrognathia combined with SBQ) was significantly improved. After the patient is admitted to the hospital, well-trained nurses can obtain the patient's WC, presence of micrognathia and SBQ score easily, and then according to the nomogram, the probability that the patient may have moderate/severe OSA can be obtained. It may be inconvenient for the physician to obtain the patient's anthropometric and oropharyngeal parameters, but it is very convenient, accurate, and efficient to obtain the above anthropometric indicators by professional nurses. The new screening mode can not only increase the screening accuracy of OSA in ACS population, but may also improve efficiency and save medical resources. After screening, moderate/severe OSA patients with ACS can be directly referred to the sleep center for PSG.
This study had several limitations. First, this was a single-center study; Second, the nurses in the cardiology ward are unable to readily identify micrognathia based on cephalometric X-ray results (sella nasion B point angle is too small), so they may only be able to identify patients with very obvious micrognathia; Third, there may be intra- and inter-observer variability because anthropometric and oropharyngeal parameters were not measured multiple times in a row on the same patient, and the same patient was not measured by different nurses. Fourth, we have not conducted validation in other cohorts at present, but this will be our next research plan.