Statistical methods
Statistical analyses were performed using the Statistical Package for the Social Sciences for Windows (SPSS version 22,0; IBM Corp, Armonk, NY, USA). The relationships between the demographic data, HEMII-pH, symptoms and sleep findings were investigated through multivariate analysis. The outcome comparison between patient groups (LPR + OSASversus LPR) was performed through Mann-Whitney U Test. A level of significance of p<0.05 was used.
Results :
A total of 89 patients completed the study. There were 45 patients with LPR and 44 subjects with both OSAS and LPR. The demographic and epidemiological data of both groups are described in Table 1. Patients with OSAS and LPR had significant lower proportions of esophagitis, hiatal hernia and gastritis compared with LPR patients. OSAS individuals had higher BMI than patients without OSAS (p=0.001). Table 2 reports the clinical presentation of LPR of both groups. The following symptom scores of LPR patients with OSAS were significantly lower than the scores of those with LPR: dysphonia; throat pain; throat clearing; excess throat mucus and ear pressure/pain. Thus, otolaryngological RSS and related quality of life (QoL) RSS were lower in patients with OSAS and LPR compared with LPR patients.
The HEMII-pH profiles of patients are available in Tables 3 and 4. Irrespective to the daytime period, patients with LPR had significant higher number of acid and nonacid HREs than patients with OSAS and LPR. Both groups were comparable regarding the nighttime HREs and the GERD features.
The PSG features of patients with both OSAS and LPR are described in Table 5. There were 12 and 13 individuals with mild (AHI: 5-15) and moderate (AHI: 16-30) OSAS, respectively. Nineteen patients had severe OSAS according to the AHI (>30). There were significant associations between ESS, BMI (rs=0.491; p=0.003), QoL RSS (rs=0.549; p=0.001) and the number of HREs in the evening (rs=0.415; p=0.015). The multivariate analysis revealed that patients with longer paradoxical sleep had higher RSS (rs=0.399, p=0.009). There was no additional significant association in the OSAS population and in subgroups of patients regarding AHI categories. There was no significant association between the severity of OSAS and the severity of LPR in the cohort. The analysis of the occurrence of arousals and hypopharyngeal reflux events revealed that only 3 patients had nighttime HREs, and, among them, the HREs were not followed by arousals/awakenings.
Subgroup analysis reported that the presence of GERD at the HEMII-pH was associated with a high number of HREs in patients with OSAS, especially nighttime (Table 6). There was a statistical trend in the association between DeMeester score and AHI (rs=0.539; p=0.070). Sleep outcomes did not differ between OSAS patients regarding the presence of GERD.