Nocturnal Hypoxemia, a Key Parameter in Overlap Syndrome

CORINA EUGENIA BUDIN1, ANCA DIANA MAIEREAN1, EDITH SIMONA IANOSI2, ADRIANA SOCACI3, ANCA DANA BUZOIANU1, TEODORA GABRIELA ALEXESCU1*, MIHAI OLTEANU4, ELENA RUSU5, COSMIN ALEC MOLDOVAN5, ROXANA MARIA NEMES5 1Iuliu Hatieganu University of Medicine and Pharmacy, 4 Louis Pasteur Str. 400349, Cluj Napoca, Romania 2University of Medicine, Pharmacy, Science and Tehnology Targu-Mures, Discipline of Pneumology, 38 Gh. Marinescu Str., 540139, Tg. Mures, Romania, 3Clinical Hospital for Infectious Diseases and Pneumology Dr. Victor Babes, 13 Gheorghe Adam Str., 300310, Timisoara, Romania 4University of Medicine and Pharmacy Craiova, 2 Petru Rares Str., 20039, Craiova, Romania 5Titu Maiorescu University, Faculty of Medicine, 67A Gh. Petrascu Str. 031593, Bucharest, Romania

Chronic Obstructive Pulmonary Disease (COPD) is a common, preventable and treatable disease that is characterized by persistent respiratory symptoms and airflow limitation that is due to airway and/or alveolar abnormalities usually caused by significant exposure to noxious particles or gases [1] and tobacco [2]. COPD has become a problem of public health that occurs in >10% of adults over 40 years of age and accounts for >5% of physician office visits and 13% of all hospitalizations, with the majority of patients remaining undiagnosed and untreated [3,4]. The burden of COPD is increasing and new guidelines focus on symptom control and the impact of comorbid conditions. Obstructive sleep apnea (OSA) is a sleep-related breathing disorder that involves a decrease or complete halt in airflow despite an ongoing effort to breathe, resulting in nocturnal hypoxemia and arousal of sleep [5]. The prevalence of OSA is increasing with age. Although it is typically a condition affecting men, with a ratio of approximately 10:1 between men and women, the prevalence of OSA is about 10% among women.However, the sex disparity of OSA ends at the age of 55 years with a sharp rise among postmenopausal women [6][7][8][9].
The first who described the overlap syndrome (OS) was Flenley and it is defined as a synergistic relation between OSA and COPD [10]. There are some data about the prevalence of overlap syndrome. COPD and OSA are both common diseases affecting 10 and respectively 5% of the adult population over 40 years of age. Their coexistence was studied in the Sleep Heart Health Study, a large community-based cohort study, including as diagnostic tests polysomnography and spirometry, in which 0.5-1% of the participants had airflow obstruction [11,12]. Patients with the overlap syndrome will present the clinical features of each disorder to a greater or lesser extent depending on the predominant disease [13]. However, there are also likely to be additional clinical features to reflect the higher prevalence of hypoxemia, hypercapnia and pulmonary hypertension. Thus, morning headaches reflecting hypercapnia, cyanosis reflecting hypoxemia and peripheral edema reflecting cor pulmonale are probably common in overlap patients [14]. Given the impact of those comorbidities in patient quality of life and morbidity, it seems logical to investigate predictors that would help identify overlap subjects. Although age, gender, neck circumference have been shown to be good predictors in some epidemiologic studies, these predictors may vary according to the specific population studied [15]. An objective anamnesis and clinical examination, including anthropometrics data, medical history andscreening questionnaires regarding symptoms may be helpful in the initial evaluation of likely OSA in COPD patients.The recent report of Ustun et al. indicates that readily available objective data from medical records such as age, sex, BMI and the presence of co-morbidities such as hypertension may be superior to the traditional symptoms of snoring and excessive daytime sleepiness in predicting the likelihood of OSA [16].

Experimental part
Sixty-two patients were enrolled in this retrospective, observational, experimental cross-sectional study. The study, conducted on human subjects, had the approval of the Medical Ethics Committee, and each subject signed an informed consent according to current legislation. All the patients were diagnosed and treated in the Leon Daniello Clinical Hospital of Pneumology Cluj-Napoca, in the Sleep Laboratory. Enrolments began in august 2016 and ended in September 2018 and inclusion criteria included: age ≥18 years old, diagnosis of OSA (mild, moderate or severe), COPD diagnosis according to GOLD criteria (GOLD I, II, III, IV). We collected anthropometric and demographic data including: age, gender, body mass index (BMI), neck circumference, apnea-hypopnea index, duration of nocturnal desaturations, the presence of associated cardiovascular diseases, defined as: diagnosis by a specialist or lack of medical documentation but patient undergoing treatment for: essential hypertension, ischemic heart disease, heart rhythm disorders, heart failure(table 1). We also achieved the Epworth Sleepiness Scale. All subjects were tested by a cardio-respiratory polygraph with 7 channels, recording nasal respiratory flow, SpO2, cardiac frequency, snoring, body position, thoracic respiratory effort and abdominal respiratory effort.
The diagnosis of COPD was established in accordance with the GOLD 2016 criteria. Diagnosis of Sleep Apnea Syndrome was performed based on a nocturnal cardio respiratorypolygraph, clinical examination, Epworth Sleepiness scale (ESS) and anamnestic data, according to AASM 2016 guidelines. Subjects were divided into two groups as follows: Group I -subjects with COPD diagnosed with OSA (OS) and group II -subjects diagnosed with OSAwithout COPD.
Data were recorded and processed using Microsoft Office Excel (Microsoft, Redmond, WA, USA). Statistical analysis was performed using IBM SPSS Statistics 17 (Statistical Package for the Social Sciences, SPSS Inc., Chicago, IL, USA). A value of p<0.05 was considered significant. The quantitative variables were reported as means± standard deviation, the ordinal variables as medians, whereas the qualitative variables were represented in absolute and relative frequency.The correlation coefficients were calculated between continuous and/or ordinal variables and their level of significance was assessed.

Results and discussions
As mentioned previously, the subjects were divided into two groups: subjects diagnosed with OS in the first group and subjects diagnosed with OSA (without COPD) in the second group. Each category included a total of 31 patients. The study population had homogeneous demographic and anthropometric characteristics. Of the 62 patients enrolled in the study, in terms of gender, the prevalence of overlap syndrome was higher in males, in both groups, 83.87% in Table 1 THE DEMOGRAPHIC AND ANTHROPOMETRIC PARAMETERS STUDIED OS group (n = 26) and 64.51% (n=20) in the OSA group. Similar results were also obtained by Soler et al. in a study conducted on a total of 44 patients, of which 54% were men [17]. Other studies have shown that males diagnosed with COPD had a higher predisposition to develop OSA [18].
On the other hand, referring to age, in the OS group the average age was 65.6 years, and in the OSA group it was 55 years (p = 0.000001), with a high statistical significance of the difference between the two groups (table 1). It is known that aging becomes an independent risk factor for OSA and COPD, data confirmed by the study by Chaouat A. et al. [19].
In the present study there was no statistically significant difference in BMI, so values in both groups were uniform. Xavier et al. have demonstrated that BMI ≥ 25 kg/m 2 is a good predictor for OSA [17]. In addition, our study demonstrates thatin the OS group the increasedBMI was associated with a higher AHI (Pearson correlation coefficient r=0.695, p<0.001) ( fig. 1). Based on the observation that a BMI greater than 25 kg/m 2 is associated with better survival, it has been proposed that obesity could be protective in COPD [20]. Also, it is plausible that the effects of air trapping on the upper airway could overcome the effects of airway collapse due to obesity or emphysema [21].Taking into consideration the neck circumference as an element that indicates central obesity, our study found a difference of 3.79 cm (48.41 cm versus 44.62 cm) (table 1) between the OS group and the OSA group, possibly due to anatomical and physiological factors and treatment with corticosteroids that increase fat deposition in the neck region. Other studies confirmed that central obesity is a risk factor for OSA in the general population [22,23]. It is known that OSA has an independent contribution to the pathogenesis of metabolic abnormalities, including type 2 diabetes, the metabolic syndrome and fatty liver disease, and a study pointed to a strong correlation between AHI, average SpO2 and HbA1c(p<0.001). It is clear that future studies are needed to better understand this relationship [24]. According to these findings, one of the most important interventions in the treatment of OSA is the nutrition program, implemented to obtain weight loss in order to reduce the apnea-hypopnea index and the severity of OSA and, inselected cases with associated craniofacial deformities,early orthodontic treatment could be beneficial [25][26][27].
Regarding OSA, the most important symptom is excessive daytime sleepiness because it can cause road and work accidents. The ESS has been used for decades as a screening tool for OSA [28]. In the Sleep Heart Health Study, an increased ESS score was associated with severity of sleep-disordered breathing (SDB), habitual snoring, male gender and other important clinical symptoms [29]. The median ESS is 14.19 in the OS group, whereas in the OSA group, the median ESS is 11.23(p=0.003) (table 1). In our study, it was a significant correlation between ESS and AHI in the OS group (Spearman correlation coefficient ρ=0.373, p=0.038) ( fig. 2), so the severity of OSA was predicted by the ESS scale in the OS patients.The patients with OS have poor sleep quality and disrupted sleep  architecture because of the respiratory symptoms such as cough or dyspnea that disrupt sleep. Sleep disturbance is one of the most common symptoms reported by COPD patients, occurring in ~40% of patients in one large study [30,31]. These patients have problems initiating or maintaining sleep, and have increased light sleep and reduced REM sleep, frequent sleep stage shifts and microarousals.Although ESS grades the severity of sleepiness,this symptom alone may be very common in COPD and, therefore lacking in predictive value [30,32].
In our study, the mean AHI in the OS group was 57.7±29.244 and in OSA group was 50.7±26.01, without any statistical significance (p=0.331) (table 1), and the same results were obtained by Radwan L. et al. [33]. In clinical practice, since overlap patients are particularly prone to oxygen desaturation during sleep, continuous overnight monitoring of arterial oxygen represents the most important variable, particularly because the recording provides data on the frequency of oxygen desaturation index (ODI). So, the ODI provides a measure of intermittent hypoxemia, which appears to be superior to the AHI in predicting cardio-vascular co-morbidity [34]. Also, COPD has been recognized to be associated with oxygen desaturation during sleep. A core element of OSA is intermittent nocturnal oxygen desaturation associated with apnea and hypopnea, and patients with overlap syndrome demonstrate even greater degrees of oxygen desaturation. Overlap patients have worse sleep-related chronic hypoxemia and hypercapnia than patients with COPD or OSA alone. OSA has a similar prevalence in COPD patients asin a general population of similar age, but oxygen desaturation during sleep is more pronounced when the two conditions coexist [12].This fact was demonstrated in our study, with a mean period of SaO . 4). Also, there are significant differences regarding the period of desaturation below 80% between the two groups, with a mean period of 60.53±80.01min in the OS group, versus 24.53±47.06min in the OSA group (p=0.235) (table 1).This means that the time spent in nocturnal hypoxemia with SpO2 lower than 80% correlates with AHIin the OS subjects(Pearson correlation coefficient r=0.599, p<0.001) (fig. 5). These positive correlations between nocturnal hypoxemia and AHI areprobably a consequence of overlap syndrome, because patients with OS have lower baseline oxygen saturation during sleep, which results in greater oxygen desaturation during apnea or hypopnea as a natural consequence of the oxyhemoglobin dissociation curve [12].
In order to confirm the nocturnal hypoxemia, we compared the minimum and the medium nocturnal SpO2 between OS group and OSA group. There were significant differences regarding the minimum SpO 2 between the two groups (p<0.001), with a mean minimum SpO 2 of 68.53±10.15% in the OS group, versus 77.7±7.02% in the OSA group. Also, there are significant differences regarding the mean nocturnal SpO 2 between the two groups, with a mean SpO 2 of 85.51±4.25% in the OS group, versus 87.83±2.91% in the OSA group.The nocturnal oxygen desaturation in overlap patients typically demonstrates a pattern of intermittent desaturation from a baseline hypoxemiaduring sleep, which contrasts with the usual pattern of intermittent desaturation from a relatively normal baseline saturation in patients with OSA alone. We observed a significant improvement regarding the daytime SpO 2 between the two groups, comparing the SpO 2 at the diagnosis and after 1-year follow-up. The mean value of daytime SpO 2 at the diagnosis was 90.9±3.59% in the OS group versus 95.9±1.57% in the OSA group (p<0.001).At 1-year follow-up the mean daytime SpO2 was 94.3±2.21% in the OS group, versus 97.1±0.9% in the OSA group (p<0.001).We conclude that the treatment with CPAP along with the treatment for COPD according to the guidelines improved the clinical status and the daytime saturation.Several mechanisms may explain these results, such as bronchodilation, reduction of respiratory load, improvement of ventilation/perfusion ratio, as well as changes in the respiratory center and in the sensitivity of chemoreceptors. The same results were found in a study that included 55 patientswhich proved significant improvements in FEV 1 , FVC, and PaCO 2 a fter 6 months of nasal CPAP therapy and specific therapy for COPD [35].
Prolonged nocturnal hypoxemia and systemic inflammation represents important contributing factors to the development of cardiovascular disease and it is notable that cardiovascular and metabolic comorbidities are recognized as a common complication of both COPD and OSA [22,36,37]. Moreover, pronounced nocturnal oxygen desaturation in patients with OSis an important factor in the development of pulmonary hypertension (PH), which is especially common in such patients [21]. In our study, OS patients experienced pulmonary hypertension more frequently than patients diagnosed with OSA (P = 0.016) (table 1), probably because of the acute rises in pulmonary artery pressure (Ppa) and pulmonary vascular resistance (PVR), coinciding with sleep induced apneas and secondary hypoxemia [38,39]. Several studies have shown that a significant proportion of OSA patients also have an elevated Ppa during the day [39-41], but still the etiology of pulmonary hypertension in these patients is uncertain. Atthe same time, we found in our study, that 64.65% of the subjects with OSpresented Heart Failure (HF) versus 16.13% of those with OSA, a ver y high statistically significant correlation (p<0.001) (table 1).
The importance of screening programs is evident, moreover because some studiesshowed that only overweight itself produce early changes in cardiovascular status of the patient by decreasing the left ventricular ejection fraction, thickening the interventricular septum, increasing left ventricular mass and implementing early diastolic dysfunction [42].Thus, one could expect that cardiovascular disease would be particularly prevalent in patients with overlap syndrome, but there are limited data on this relationship.
Regarding the treatment of overlap syndrome, the most effective and preferred treatment is the therapy with continuous positive airway pressure(CPAP), along with the specific treatment for COPD according to the guidelines. The role of CPAP is to reduce the AHI, to correct the desaturations and to improve the quality of sleep. In our study, we observed that there are significant benefits of the CPAP therapy in the two groups, with a mean value of residual IAH of 3.62±2.27 in the OS group, versus 2.36±1.26 in the OSA group (p=0.01) (table 1).
Treatment of overlap syndrome is targeted at each disease state as per guidelines. Patients with overlap syndrome who are efficiently treated with CPAP will have adequate nocturnal oxygen saturation, but the therapeutic success is primarily ensured by the patients'compliance [43]. In addition, there are other concerns that must be taken into account once the CPAP therapy is initiated at home, like microclimate and the hygiene of CPAP device, including masks and tubing [44][45][46]. Some studies demonstrated the presence of various microbiological and inorganic structures on the inner surface of CPAP masks and tubes, emphasizing the risk of microbial and inorganic elements inhalation into the upper and lower airways. Once they arrive in the respiratory way, in patients with COPD these microorganisms can reduce the compliance to CPAP therapy and moreover can produce exacerbations of the respiratory symptomatology [45,46] with a high risk of hospitalization and other infections [47][48][49].

Conclusions
Our study is important in order to evaluate the patients with overlap syndrome in a complex way and to understand that patients with COPD have OSA predictors that differ from those usually recognized in the general population. The most important element in diagnosis and follow-up for patients with OS is the total nocturnal time of desaturations, because, as our study shows, the subjects with OS have more nocturnal desaturations, with a negative impact on symptomatology, cardiovascular comorbidities and management.The mainstay of treatment for patients diagnosed with overlap syndrome is continuous positive airway pressure, besides the treatment for COPD according to guidelines. It is important that clinicians should be trained that the management of selected obese COPD patientsought to include comprehensive sleep history, sleep hygiene and OSA screening, because in most of the cases, in clinical practice OSA is underdiagnosed in these patients.