Introduction

Sleep apnea syndrome (SAS) is a common chronic sleep-related breathing disorder, affecting approximately 5% of middle-aged adults, predominantly men [1, 2]. Currently considered a potentially dangerous disease, SAS poses a growing healthcare concern, increasing economic and social burdens [3]. A well-established link exists between SAS and cardiovascular morbidity and mortality, as it is implicated in a range of cardiovascular conditions, including heart failure (HF), coronary artery disease, and stroke. Additionally, SAS is associated with cardiac rhythm and conduction disorders and a high risk of sudden cardiac death during sleep [4, 5]. SAS includes a wide range of sleep-disordered breathing conditions, from central to mixed apnea, as well as obstructive apnea and hypopnea [6]. Both central and obstructive sleep apneas (OSAs) are frequently observed, with OSA occurring in 17% to 61% of cases and central sleep apnea (CSA) in 35% to 39% [7]. Polysomnography remains the gold standard for diagnosing SAS. Based on the apnea-hypopnea index (AHI), SAS severity is classified into mild, moderate, and severe forms (AHI 5–14, 15–29, and ≥ 30 events/hour, respectively) [3].

Several studies have highlighted an increased incidence of cardiac structural and functional alterations in SAS patients, particularly left ventricular hypertrophy and diastolic dysfunction [8, 9]. However, the effects of SAS on the right ventricle (RV) remain a subject of debate.

The study aimed to identify RV echocardiography parameters that predict severe SAS in patients with stable chronic HF and concomitant SAS.

Materials and methods

Results

We enrolled 85 patients from our outpatient department for stable HF.

A total of 71 patients (83.5%) were diagnosed with SAS. OSA was observed in 50 patients (58.8%) and CSA in 21 patients (24.7%). Mild SAS was more frequent. It was observed in 32 patients (37.6%), moderate SAS in 17 patients (20%), and severe SAS in 22 patients (25.8%). The characteristics of patients with SAS were summarized in Table 1.

Smoking and diabetes were the most common cardiovascular risk factors, present in 63.4% (n = 45) and 60.6% (n = 43) of patients. Eight patients (11.3%) had chronic obstructive pulmonary disease.

Medication included beta-blockers in 81.7% (58 patients), diuretics in 62.0% (44 patients), ACE inhibitors in 59.2% (42 patients), spironolactone in 53.5% (38 patients), sodium-glucose cotransporter-2 inhibitors in 42.3% (30 patients) and angiotensin II receptor blockers in 11.3% (8 patients).

The etiology of HF in our study population was ischemic in 59.2% (42 patients), idiopathic in 16.9% (12 patients), arrhythmias in 8.5% (6 patients), hypertensive in 9.9% (7 patients), and others in 5.6% (4 patients).

For the ultrasound parameters, the median value of LVEF was 38% (33; 52). Within this population, 49.3% of patients exhibited reduced ejection fraction, 14.1% had mildly reduced ejection fraction, and 36.6% had preserved ejection fraction.

The comparison of clinical and echocardiographic data between Group A (22 patients: 31%) and Group B (49 patients: 69%) was reported in Tables 1 and 2.

There were no significant differences between the two groups regarding cardiovascular risk factors, HF phenotype, or left-heart ultrasound parameters, including end-diastolic diameter, indexed left ventricular mass, LVEF, global longitudinal strain, indexed left atrial volume, and mean E/e’ ratio (all p > 0.05). However, RV FAC (p = 0.005), RVFWLS (p = 0.008), RV Tei index (p = 0.012), and right atrial surface (p = 0.027) were significantly higher in severe sleep apnea. Systolic pulmonary arterial pressure, RV diameter, and TAPSE were similar between the two groups.

To further investigate the relationship between right heart ultrasound parameters and severe SAS, ROC curve analysis was performed to determine the cut-off values of echocardiographic parameters associated with severe SAS, as observed in the above comparison.

Univariate analysis revealed that RV FAC < 40 [air under the curve (AUC) = 0.693, p = 0.010] increased the likelihood of belonging to Group A by 4 (OR = 4; 95% CI = 1.38–11; p = 0.09). For RVFWLS, a cut-off value > –17 (AUC = 0.699, p = 0.008) increased the likelihood of belonging to Group A by 3.5 (OR = 3.5; 95% CI = 1.15–9.41; p = 0.002). An RV Tei index > 0.50 (AUC = 0.692, p = 0.010) increased the likelihood of being classified in Group A by a factor of 3 (OR = 3; 95% CI = 1.06–8.56; p = 0.035).

A multivariate study identified RVFWLS [hazard ratio (HR) = 0.892; 95% CI = 0.816–0.979; p = 0.036] as an independent predictor of severe SAS after binary logistic regression analysis, as shown in Table 3. A one-point decrease in RVFWLS was associated with an 11% increase in the risk of developing severe SAS.

Discussion

This cross-sectional study included 71 patients (83.5%) diagnosed with chronic HF and SAS.

The study revealed a notably high prevalence (83.5%) of SAS in HF patients, underscoring the significant association between these two major public health concerns. This finding raises the question of whether routine screening for SAS should be implemented in HF patients. According to the literature, the prevalence of SAS in HF patients ranges from 46% to 85% [12–17]. Using an AHI threshold of ≥ 5/h, several studies have reported that the prevalence of SAS in HF patients varies between 60% and 85% [14–19]. Using an AHI threshold of ≥ 5/h, several studies have reported that the prevalence of SAS in HF patients varies between 60 and 85% [16, 19–22]. The elevated frequency of SAS in our population can be explained by the high frequency of patients with HF with reduced and moderately reduced ejection fractions.

In patients with chronic HF, the study found that while RVFWLS, RV Tei index, and RV FAC were associated with severe SAS in univariate analysis, only RVFWLS, with a threshold of –17, emerged as an independent predictor of severe SAS in HF patients with SAS. A 1% decrease in RVFWLS was associated with an 11% increase in the risk of developing severe SAS (HR = 0.892; 95% CI = 0.816–0.979; p = 0.036).

Several studies have examined right ventricular remodeling and dysfunction in individuals with SAS, exploring the role of RV echocardiographic parameters in predicting the severity of SAS [23]. However, our study is the first to specifically investigate these parameters in patients with chronic HF, addressing a significant gap in the literature. Notably, there is a paucity of research on this topic within the context of HF, and our study contributes valuable insights in this area.

RV strain is highlighted as a more sensitive early imaging marker for assessing the longitudinal function of the RV compared to conventional echocardiographic parameters commonly used in estimating global and regional RV systolic function.

In a study by Tadic et al. [23], significant impairment in RVFWLS was observed in a large number of patients with obstructive SAS, demonstrating a progressive decline with increasing severity of SAS. RVFWLS exhibited more negative values (indicating worse strain) in patients with moderate SAS compared to those with mild SAS [standardized mean difference (SMD): 1.29 ± 0.21, 95% CI: 0.87–1.70, p < 0.0001, data from three studies]. Similarly, patients with severe SAS showed more negative RVFWLS values compared to those with mild SAS (SMD: 1.26 ± 0.20, 95% CI: 0.87–1.65, p < 0.0001, and data from three studies).

However, this study did not report a statistically significant difference between moderate and severe SAS. Notably, our study is unique in including patients with HF, a population that was excluded from the previous investigation.

Consistent with these findings, Hammerstingl et al. [9] reported a lower RV global longitudinal strain in subjects with higher AHI, showing a positive correlation with SAS severity. Tugcu et al. [5] also found significant correlations between AHI and various RV parameters, and RV mid-free wall strain and strain rate demonstrated the strongest associations. In our study, we specifically focused on RV-free wall strain to exclude the influence of the left ventricle. Notably, after performing binary logistic regression analysis, RVFWLS (HR = 0.892; 95% CI = 0.816–0.979; p = 0.036) emerged as an independent predictor for belonging to the severe SAS Group.

RV FAC provides an estimate of overall RV systolic function. In a study by Romero-Corral et al. [24], involving 85 cases of obstructive SAS, a severe decline in RV FAC was observed in the severe SAS group compared to the mild SAS Group (53 ± 7% vs. 57 ± 6%; p = 0.01). Consistent with these findings, our investigation revealed a significant association between RV FAC and severe SAS in the univariate analysis. An RV FAC (%) below 40 increased the likelihood of being classified as having severe SAS by a factor of 4 (OR = 4; 95% CI = 1.38–11; p = 0.09).

In a previous study conducted by Dursunoglu et al. [25], global right ventricular function was assessed using the RV Tei index in individuals diagnosed with SAS, excluding those with a history of cardiac or lung diseases. Their research revealed a robust correlation between the RV Tei index and AHI (r = 0.84; p < 0.001), which closely aligns with the findings in our study.

More recently, Akyol et al. [26] established a proportional relationship between the RV Tei index and the severity of obstructive SAS. Some studies have shown that continuous positive airway pressure therapy can improve the RV Tei index. Specifically, RV Tei index measurements taken at baseline and after 6 months of continuous positive airway pressure therapy demonstrated significant improvement in patients with OSA who did not have hypertension [25]. Romero-Corral et al. [24] highlighted altered right and left ventricular function in patients with obstructive SAS, particularly in those with moderate to severe SAS.

In our study, the RV Tei index, with a cutoff value > 0.50 (OR = 3; 95% CI = 1.06–8.56; p = 0.035), also demonstrated a correlation with the severity of SAS, though this was observed only in univariate analysis.

The impact of obstructive SAS on RV systolic function (RV S’) remains controversial. Altekin et al. [27] in a study involving 79 patients, found no significant variation in RV S’ across different severity grades of obstructive SAS, consistent with our findings. However, Shivalkar et al. [28] reported a significant correlation between AHI and RV S’.

In a study by Ibn Hadj Amor et al. [3] assessing right ventricular remodeling and dysfunction in obstructive SAS, TAPSE did not show a significant correlation with SAS severity. Similarly, Tadic et al. [23] observed a reduction in TAPSE in patients with SAS compared to controls, but this parameter proved insufficient for differentiating between SAS severity grades. Our study corroborates these findings.

To our knowledge, this is the first study to assess the role of RV echocardiographic parameters in predicting the severity of SAS in patients with both chronic HF and SAS. However, this study has limitations. It was a single-center study with a small number of patients, which limits the generalization of the results. Further multicenter studies with larger populations are required to confirm our findings.

In conclusion, we found that RVFWLS emerged as the only reliable indicator of SAS severity in individuals with both SAS and HF, underscoring its significant role in assessing SAS severity in this patient group. Therefore, a thorough assessment of RV function is necessary in HF patients with SAS, as it facilitates the detection of those with severe SAS.